Performance of the equations in men and women in QResearch validation cohort and CPRD validation cohort
| Statistic | QResearch validation cohort Mean (95% CI) | CPRD validation cohort Mean (95% CI) |
|---|---|---|
| Women | ||
| D statistic | 1.66 (1.60 to 1.72) | 1.71 (1.66 to 1.76) |
| R2 | 39.77 (38.07 to 41.46) | 41.15 (39.72 to 42.58) |
| ROC | 0.770 (0.762 to 0.778) | 0.783 (0.776 to 0.789) |
| Men | ||
| D statistic | 1.67 (1.62 to 1.72) | 1.63 (1.58 to 1.67) |
| R2 | 40.04 (38.54 to 41.53) | 38.72 (37.44 to 40.00) |
| ROC | 0.764 (0.757 to 0.771) | 0.769 (0.763 to 0.775) |
Discrimination is the ability of the risk prediction model to differentiate between patients who experience an admission event during the study and those who do not. This measure is quantified by calculating the area under the ROC statistic; where a value of 0.5 represents chance and 1 represents perfect discrimination.
The D statistic is also a measure of discrimination which is specific to censored survival data. As with the ROC, higher values indicate better discrimination. R2 measures explained variation and higher values indicate more variation is explained.
ROC, receiver operating characteristic curve.